bobinator said:

---

I think we've all been sleeping on Wisconsin.

---

I wouldn't say sleeping. We have been acknowledging them. We may be undervaluing them slightly. I think their ceiling is a 2 seed. But I ultimately think they will end up in the 3 or 4 seed slot.

Honestly given the schedule I wouldn't call 3-3 the last 6 games a total collapse and I think that finish probably puts us as a 3 seed.

Just getting a split this week will be a good accomplishment and we still have Florida and auburn after that.

Wisconsin is absolutely in play for a 1, but will need some help to get there. They only have 2 more quad 1 games left. Their win over Arizona in non conference is key but they have to win out to have a shot and hope that Florida/A&M/Houston all drop a couple. Probably Iowa State too.

Yeah but I don't think any of us were really looking at them as a 1/2 seed threat until now. Maybe I shouldn't speak for everyone but their resume kind of caught me off guard last night and it looks like Colleyville kind of felt the same way.

Their resume numbers are already a lot better than I expected.

Do Conference Tournament games mean much at this point?

Bupkis said:

---

Do Conference Tournament games mean much at this point?

---

Only for bid thieves.

OP Updated

New Locks: Mississippi State, Marquette
Eliminations: Arizona State, Villanova

Moving Up: Wisconsin, Mississippi State, BYU, Michigan State
Moving Down: Kansas, Oklahoma, UCLA

That game was worse for our numbers than I'd have liked. Losing isn't a big deal, I actually had that one pegged as a loss, but getting killed was a bad result.

On the plus side we still have the 7th best resume numbers and a pretty decent sized gap before Iowa state, Michigan state, Purdue and Mississippi state. Predictives were always a bit of a worry, just puts more pressure to perform well this weekend vs. Tennessee who is very high in the metrics.

I was about to post that same thing - big drops in NET, Kenpom, Bart Torvik (who's model really hates us) and BPI.

To me, thats the frustrating thing about all those predictive metrics, they can swing wildly even this late in the season when you play poorly and get the doors blown off, even on the road. It rewards and punishes a style of play and late game strategies, encouraging teams to either keep the foot on the gas even when there is no doubt, or try to scramble and make the score "more respectable" late even though you have no chance.

Had the game been 64-54 or 70-61 instead of 70-54 due to a couple late baskets (or missed baskets), the numbers don't change nearly as much, but was the game any less or an a$$ kicking? No.

I just have a hard time understanding the math as to how this late in the season (with all the data already accrued), one game can swing numbers so much.

Well KenPom and Torvik both use kind of a diminishing trail where the more recent a game is the more it matters in the numbers. Which makes sense, you want an idea of how good a team is right now, recent games matter more. The NET doesn't do that, but that was still a brutal game for our offense, so not surprised it moved the numbers quite a bit.

But like Colleyville said, the resume metrics are still good, and that's what really matters.

Torvik's model probably hates us more because of the game control factor. We've won at least two games this year where we didn't control the game at all (Ole Miss and Oklahoma) and we just don't bury opponents the way some other teams do. Though we did also lose a game that we controlled most of the way (Texas.) Like Houston, for example, the reason they're so dominant in the metrics every year, other than being very good, is that they just absolutely hammer teams that aren't on the same level as they are.

Because of our offense, we just don't do that.

bobinator said:

---

Well KenPom and Torvik both use kind of a diminishing trail where the more recent a game is the more it matters in the numbers. Which makes sense, you want an idea of how good a team is right now, recent games matter more.

---

Yep, that does make sense.

The raw numbers are still pretty bunched up where we are so it doesn't necessarily even take wild swings to move 5-6 places.

In BPI we are tied with the 2 teams above us and it's close enough that if UCF beats Oklahoma State something like that may move us up 2 spots.

In torvik we are tied at the 10,000th decimal place with St. John's. Who is ranked above us and very close to the next 7. It is still very tight in some spots.

Yeah that's a good point too.

This weekend would be a prime weekend for some old Big 12-style bid maximization strategy.

State, Ole Miss, Auburn and Missouri should all throw their games.

bobinator said:

---

Torvik's model probably hates us more because of the game control factor. We've won at least two games this year where we didn't control the game at all (Ole Miss and Oklahoma) and we just don't bury opponents the way some other teams do. Though we did also lose a game that we controlled most of the way (Texas.) Like Houston, for example, the reason they're so dominant in the metrics every year, other than being very good, is that they just absolutely hammer teams that aren't on the same level as they are.

Because of our offense, we just don't do that.

---

Torvik's model also includes a luck factor and a momentum factor (separate from the recency weighting). It's a thousand times better than Pomeroy's model which assumes that all efficiency stats happen in a vacuum

bobinator said:

---

This weekend would be a prime weekend for some old Big 12-style bid maximization strategy.

State, Ole Miss, Auburn and Missouri should all throw their games.

---

I kept waiting for the Big 12 Kentucky quip last Texas game given the Horns were on life support.

While you can never be unhappy about making the Dance, I'd be fine if Texas lost the remainder of its games and bundled Terry off into the sunset with a large bag.

OP Updated through Friday's action.

Locked Up: St. John's, St. Mary's
Eliminated: Washington

Moving Up: Missouri, Michigan State
Moving Down: Marquette, UC Irvine